/distracting_feature

This is a PyTorch implement of ''Abstract Reasoning with Distracting Features'' that appears in the NeurIPS 2019.

Primary LanguagePython

Abstract Reasoning with Distracting Features

To appear in NeurIPS 2019.

Abstract Reasoning with Distracting Features

(Kecheng Zheng, Wei Wei and Zheng-jun Zha, 2019)



Dataset

To download the dataset, please check chizhang's project page.

Performance

For details, please check our paper.

Dependencies

Important

  • Python 2.7
  • PyTorch
  • CUDA and cuDNN

Usage

Benchmarking

python main.py --net <model name> --datapath <path to the dataset> --rl False --typeloss False

#TODO

  1. The code for PGM dataset
  2. The long-term features (state) in teacher model

Citation

If you find the paper and/or the code helpful, please cite us.

@inproceedings{zheng2019abstract,
    title={Abstract Reasoning with Distracting Features},
    author={Kecheng Zheng and Zheng-jun Zha and Wei Wei},
    booktitle={Advances in Neural Information Processing Systems},
    year={2019}
}

Acknowledgement

Contact

Please feel free to discuss paper/code through issues or emails.

License

Apache License 2.0